|Publication number||US7668966 B2|
|Application number||US 10/286,576|
|Publication date||Feb 23, 2010|
|Filing date||Nov 1, 2002|
|Priority date||Nov 2, 2001|
|Also published as||EP1449106A1, US20030088529, WO2003040947A1, WO2003040947A9|
|Publication number||10286576, 286576, US 7668966 B2, US 7668966B2, US-B2-7668966, US7668966 B2, US7668966B2|
|Inventors||Eric Klinker, Jeremy Johnson, Steven McCanne, Daniel S. McKernan, Gordon Chaffee, Charles J. Fraleigh|
|Original Assignee||Internap Network Services Corporation|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (103), Non-Patent Citations (11), Referenced by (32), Classifications (34), Legal Events (9)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This application is related to the below-referenced U.S. Provisional and Non-Provisional Patent Applications. The subject matter of the related applications is hereby incorporated by reference. The related applications are commonly assigned.
The present invention relates generally to the control of data over network communications systems, and more specifically to a data network controller.
Data networks are communications systems that connect nodes for the purpose of sharing computing resources. Creating efficient data routes between source and destination nodes is an important goal of organizations using data networks to facilitate the sharing of resources, data, and information.
A node is typically represented by one or more computing devices, which can include personal computers and peripheral equipment. Interchangeably referred to herein as a “point,” a node is typically an endpoint for a particular segment on a data network known as either a “route” or “path.” The segment defined between two consecutive nodes is hereinafter referred to as a “hop.” Additionally, an autonomous system (“AS”) hop refers to a segment within, for example, a single autonomous system. Further, a hop can also be a portion of a network path that is defined by two non-consecutive nodes.
For example, 15 individual nodes may exist between a source and a destination. These 15 nodes may span three autonomous systems, thus a total of 7 hops might be defined. The first hop could run from an egress node of the source network to an ingress node of the first AS. The second hop could run across the first AS between its egress and ingress nodes. The third hop could run from the egress node of the first AS to an ingress node of an second AS. The fourth hop could run from an ingress node to an egress node of the second AS. The fifth hop could run across an egress node of the second AS to an ingress node of the third AS. The sixth hop could run between an ingress node and an egress node of the third AS. Finally, the seventh hop could run between an egress node of the third AS to the destination node. Although the above example listed seven hops, a data path such as that described above, may be composed of fewer or greater than the number of hops described. Moreover, more than one hop could run between the ingress and egress nodes of an AS.
A source and/or a destination may embody a data source, such as a web-based database, storage enclosure, series of storage facilities, or the like that maintains computational resources for a particular organization. Resources may include mission-critical data or software programs, or applications, which are used to execute specific missions. For example, in a banking context, a database may contain sensitive and secure financial information for clients or partners. This information is often required to support mission-critical applications. In many situations where large distributed organizations depend upon the sharing of such resources, data networks are implemented and often through the use of more than one network service provider or “NSP.” Where more than one NSP is used for a network, the network is said to be “multi-homed.” Where a multi-homed network exists, there are several drawbacks to conventional data network control.
First, efficiency of routing data is a significant problem with regard to conventional data network control systems. Generally, efficiency includes optimizing network performance and minimizing costs. Reducing costs and maximizing performance (e.g., reducing loss, latency, jitter, etc.) are common goals among conventional data routing control techniques and technologies. However, conventional data network control systems are unable to achieve maximum performance at minimum cost for most destinations. Efficiency is often sacrificed when considered against the cost of data transmission and performance of a data network. In other words, cost and performance are factors which often, either implicitly or explicitly, override efficiency. Although efficiency is often critical, it is difficult to achieve because there are numerous factors influencing the logic and decision-making processes in terms of how, when, and where to route data between a source and a destination. Multi-homed networks have inherent barriers to achieving efficient data routing due to the use of multiple NSPs. Thus, multi-homing requires the evaluation of too many routes for determining candidate data paths with conventional route control techniques.
Protocols such as Border Gateway Protocol (“BGP”) is a data protocol used to standard data communications between ASes. For multi-homed enterprises or organizations, BGP is a useful protocol. However, in terms of efficiency, BGP may not necessarily provide the optimum solution. BGP does not make routing control decisions based upon either cost information or performance statistics. Instead, BGP makes decisions on AS paths and other administrative settings. And even though BGP distributes data load on NSPs, it does not distribute data loads to minimize bandwidth usage costs.
Further, BGP routes data to a destination in only one way such that performance is not considered in routing control decisions by traditional use of BGP. Although a better performing route may exist, BGP will not shift data traffic to the better performing route. In another example, an optimum performing route may exist, but the route may later degrade. Since BGP is unable to account for the subsequent degradation, if cannot adjust to a better performing route to resolve such degradation. As an example, a shorter, lower performance route and a longer, higher performance route might both exist to the same destination. BGP will make routing control decisions to direct data traffic flow over the shorter route. Although the longer route is the better performing route, BGP and conventional protocols will select the shorter route, despite the lower-performance quality of the selected route. Therefore, there is a need for a solution to control data routing while maximizing performance and minimizing costs for most destinations.
Another drawback to conventional data network controllers is usage overruns. NSPs can set usage thresholds by generally measuring flow volumes or bandwidth. These thresholds define maximum utilization rates and, for example, are often user-configurable. When maximum utilizations rates are met, any excess data flow spills over into another, generally more expensive NSP usage profile such that the excess data flows cost more to route. Inflexible provider pricing tiers result in incrementally more expensive buckets in which overflow data is placed. Rather than re-allocate bandwidth needs in a more efficient manner, conventional data route control techniques often shift maximum and/or minimum pricing limits, resulting in dynamic and static pricing models. Ultimately, the multi-homed enterprise or organization is not maximizing the efficiency of its data routing requirements and, thus, pay additional costs unnecessarily. In large distributed networks, these costs may result in significant financial payments far beyond what is required given the amount of data that needs to be routed to support any mission-critical applications. For example, a significant part of these costs is the time-instantaneous cost of using a given NSP.
Further, many traditional route control products and methods today utilize and make control decisions on routes as advertised in a protocol routing table, such as a Border Gateway Protocol (BGP) routing table. These routes are often large allocations of address space meant to keep the inter-provider routing table small. Unfortunately, these route control products do not take into consideration the vast geographic distances between adjacent networks of a large address project. Geography can affect the performance of routing data because one or more paths can degrade due to, for example, congestion caused by network outages, line cuts, etc. Therefore, a controller decision intended to correct a routing problem in a network, especially in a multi-homed situation, could have far-reaching and adverse effects upon numerous addresses, depending upon the number of addresses affected.
Therefore, what is needed is a system and method for overcoming the aforementioned drawbacks of conventional route controllers and route control techniques.
The present invention is directed to a data network controller. The data network controller comprises a control module for controlling a network. The data network controller also analyzes aggregate flow information via an interface for communicating aggregate flow information to other modules. Information received by a cache is shared with data network controller and elements of a flow control system. An active calibrator communicates network data to the data network controller and an aggregation module is used for aggregating flow information. A bus, such as a software bus, is used for inter-module communication. A user module provides user information, which is communicating to the control module. A file library stores data for provider information and communicates the information to the control module. Additional information such as traffic and policy information is stored in a repository which can provide request information to the data network controller.
The present invention is directed to a data network controller. The data network controller comprises a control module for controlling a data network. The data network controller also analyzes aggregate flow information via an interface for communicating aggregate flow information to other modules. Information received by a cache is shared with data network controller and elements of a flow control system. An active calibrator communicates network data to the data network controller and an aggregation module is used for aggregating flow information. A bus, such as a software bus, is used for inter-module communication. A user module provides user information, which is communicating to the control module. A file library stores data for provider information and communicates the information to the control module. Additional information such as traffic and policy information is stored in a repository which can provide request information to the data network controller.
Another embodiment of the present invention discloses a method for controlling a network. The method comprises evaluating a candidate latency, evaluating a candidate loss, comparing a candidate bandwidth to a bandwidth allocation, determining a loss value, determining a cost, evaluating an identifier, and determining a change value.
The method further comprises determining a lowest latency if the candidate bandwidth is less than the bandwidth allocation, determining a lowest loss, determining a greatest available bandwidth based on a cost and a usage variable, determining a cost tier, determining a lowest identifier, and assigning a change value.
In another embodiment of the present invention a method for controlling a network is disclosed. The method comprises comparing a candidate loss to a current loss, comparing a candidate latency to a current latency, determining a best latency, determining a best loss, evaluating a candidate bandwidth value, determining a maximum candidate bandwidth value, determining a cost tier, determining a lowest identifier, and assigning a change value.
Other embodiments of the present invention disclose an apparatus for controlling a network. The apparatus comprises a means for evaluating a candidate latency, a means for evaluating a candidate loss, a means for comparing a candidate bandwidth to a bandwidth allocation, a means for determining a loss value, a means for determining a cost, a means for evaluating an identifier, and a means for determining a change value.
The apparatus further discloses a means for determining a lowest latency if the candidate bandwidth is less than the bandwidth allocation, a means for determining a lowest loss, a means for determining a greatest available bandwidth based on a cost and a usage variable, a means for determining a cost tier, a means for determining a lowest identifier, and a means for assigning a change value.
Another embodiment of the present invention discloses an apparatus for controlling a network where the apparatus comprises a means for comparing a candidate loss to a current loss, a means for comparing a candidate latency to a current latency, a means for determining a best latency, a means for determining a best loss, a means for evaluating a candidate bandwidth values, a means for determining a maximum candidate bandwidth value, a means for determining a cost tier, a means for determining a lowest identifier, and a means for assigning a change value.
Still another embodiment of the present invention provides for a computer-readable medium storing instructions for controlling a network, by performing the steps of evaluating a candidate latency, evaluating a candidate loss, comparing a candidate bandwidth to a bandwidth allocation, determining a loss value, determining a cost, evaluating an identifier, and determining a change value.
The computer-readable medium further comprises the steps of determining a lowest latency if the candidate bandwidth is less than the bandwidth allocation, determining a lowest loss, determining a greatest available bandwidth based on a cost and a usage variable, determining a cost tier, determining a lowest identifier, and assigning a change value.
In another embodiment of the present invention, a computer-readable medium storing instructions for controlling a network is disclosed. The computer readable medium performs the steps of comparing a candidate loss to a current loss, comparing a candidate latency to a current latency, determining a best latency, determining a best loss, evaluating a candidate bandwidth values, determining a maximum candidate bandwidth value, determining a cost tier, determining a lowest identifier, and assigning a change value.
Another embodiment of the present invention discloses a computer data signal embodied in a carrier wave. The computer data signal comprises program code for evaluating a candidate latency, program code for evaluating a candidate loss, program code for comparing a candidate bandwidth to a bandwidth allocation, program code for determining a loss value, program code for determining cost, program code for evaluating an identifier, and program code for determining a change value.
The computer data signal embodied in a carrier wave further comprises program code for determining a lowest latency if the candidate bandwidth is less than the bandwidth allocation, program code for determining a lowest loss, program code for determining a greatest available bandwidth based on a cost and a usage variable, program code for determining a cost tier, program code for determining a lowest identifier and program code for assigning a change value.
Another embodiment of the present invention discloses a computer data signal embodied in a carrier wave comprising program code for comparing a candidate loss to a current loss, program code for comparing a candidate latency to a current latency, program code for determining a best latency, program code for determining a best loss, program code for evaluating candidate bandwidth values, program code for determining a maximum candidate bandwidth value, program code for determining a cost tier, program code for determining a lowest identifier, and program code for assigning a change value.
Detailed descriptions of specific embodiments of the present invention are provided herein. It is to be understood, however, that the present invention may be embodied in various forms. Therefore, specific details disclosed herein are not to be interpreted as limiting, but rather as a basis for the claims and as a representative basis for teaching one skilled in the art to employ the present invention in virtually any appropriately detailed system, structure, method, process or manner. Although numerous embodiments are discussed herein, the present invention is not limited to those embodiments. In many instances, for example, routing control protocols such as BGP, may be foreseeably replaced or superseded by other types of protocols for either addressing or routing. The detailed descriptions presented herein are provided to enable one skilled in the art to implement the present invention.
The present invention relates generally to routing of data over networked communication systems, and more specifically to a system and method for route control over data networks. Data transmission in a network uses protocols to standardize sharing of resources between communications devices. BGP is a protocol used between autonomous networks, particularly multi-homed networks or networks that rely upon more than one service provider to access the Internet. In many instances, data routing relies upon the use of protocols such as BGP, which can also be classified as internal and external, depending upon how BGP information is distributed among external routers. Other examples of protocols include Exterior Gateway Protocol, and Inter-Domain Routing Protocol, Stream Transmission Control Protocol (SCTP), and Transmission Control Protocol/Internet Protocol (TCP/IP) all of which standardize communication among or over data networks. Embodiments of the present invention improves efficiency in data routing control, enabling best routing control performance at the lowest network cost for all destinations. As discussed above, conventional data routing control techniques are unable to improve efficiency and minimize costs for destinations. Embodiments of the present invention leverage diverse paths and diverse billing structures and contracts to maximize efficiency. Conventional solutions can only minimize cost for some destinations, and performance for others. Performance metrics such as RTT, packet loss, and jitter can be optimized for all destinations, in accordance with embodiments of the present invention. Maximizing performance and minimizing costs in the present invention generally involve selecting the best candidate data path relative to the cost and performance of the selected path.
A data path as described herein can be a route between a first point (e.g., a source node) to a second point (e.g., a destination node), and is divided into segments or “hops,” each of which connects two consecutive nodes along a data path between a source and a destination point. Hops can also be defined as segments spanning a single autonomous system (“AS”) or connecting two or more AS. To identify a particular node in a network, typically IP or other protocol-bound addresses. A typical IP address is composed of 32-bits, composing 4 individual address sets of 8 bits each. The netmask or prefix refers to the number of relevant bits in an address. For example, an IP address of 239.051.0.0/16, in a classless interdomain routing (CIDR) schema, identifies the relevant address of a particular node. The /16 prefix identifies a netmask of 16 bits where the first 16 bits of the address are relevant for announcing or advertising this address, in accordance with one embodiment of the present invention.
Conventional data control techniques generally employ probing in order to define the network topology and assert control over data routing. In accordance with an embodiment of the present invention, data control uses prefix lengths that exist in an Internet routing table, typically identified using length designations such as “/16.” The longer the prefix, the more specific the address. A /16 prefix represents a larger subset of addresses than a /24 prefix. If a problem occurs with a point addressed with a /16 prefix, for example, a larger number of addresses would be affected than a point with a prefix of /24. Thus, a /16 prefix would affect more destinations or addresses than a /24 prefix. Conventional techniques assign arbitrary-length prefixes to route announcements. By creating topological network maps, the present invention as described below can provide, in accordance with one embodiment, for data routing control to occur with greater emphasis on efficiency and, incidentally, using fewer announcements with greater specificity. Thus, data traffic volume is reduced and the efficiency of data routing and control is improved.
Although the term “prefix” is used herein to describe the subdivision of IP addresses, it is noteworthy that the embodiments are not limited to the use of a prefix. Rather, any suitable “address set” can be substituted for “prefix,” “sub-prefix,” etc. to describe how an address of interest (i.e., destination) can be categorized. The addresses need not be contiguous in a prefix boundary and can be as small as a single active address (i.e., “/32”). Path tracing probes are sent to obtain candidate path measurements to network points up to, and sometimes including the destination, or to any network point that cannot be probed, due to disabling measures such as firewalls or filters. Identifiable convergence points, or points where more than one data route converge, are actively probed and announced for data routing. The following figures and discussion further illuminate embodiments of the present invention.
As with the external physical configuration shown in
The exemplary use of the Internet or another data network for distribution or communication of information is not strictly necessary to practice the present invention but rather is merely used to illustrate a specific embodiment. Further, the use of server computers and the designation of server and client machines are not crucial to an implementation of the present invention. USER1 Computer can be connected directly to the Internet. Server1's connection to the Internet is typically by a relatively high bandwidth transmission medium such as a T1 line, a T3 line, Metro Area Ethernet, or the like, although it might be connected in a similar fashion as with USER1. Similarly, other computers 84 are shown utilizing a local network (e.g., Local Area Network, or LAN) at a different location from USER1 Computer. The computers at 84 are coupled via Server2 to the Internet. Although computers 84 are shown to include only a single server (e.g., Server2), two or more servers can be connected to the local network associated with computers 84. The USER3 and Server3 configuration represent yet a third network of computing devices.
Although flow control system 200 is shown external of and in communication with source network 82 elements such as configuration element 208, flow control system 200 can be wholly embodied in any of the elements shown, or alternatively, can be distributed, in portions, over each of the elements such as servers 86. In another embodiment, flow control system 200 resides on one or more servers or network elements within exemplary source network 82.
An exemplary data network includes one or more source networks 82. A source network 82 typically is a local network including one or more servers owned and operated by web services providers, application/internet/network service providers, managed service providers, content delivery networks, web hosting companies, individual enterprises, corporations, entities and the like. Such service providers typically communicate information to users that are further removed from the multi-homed network service providers 82, such as NSP 1, NSP 2, NSP 3, . . . and NSPn, as illustrated in
In an embodiment of flow control system 200 operates to measure end-to-end (i.e., source to destination and destination to source) data traffic 218 in terms of flow characteristics, such as performance, cost, bandwidth, and the like. Flow control system 200 also generates statistics associated with data paths across multiple NSPs in real time, or near-real time. Such statistics are communicated to source network 82 for providing network engineering and administration personnel, for example, with report information such that on-the-fly reports can be created to provide information related to route-change activity, traffic performance as delivered to selected destinations and transit provider usage (i.e., bandwidth), cost, and the like.
In one embodiment of the present invention, a local computing device uses information extracted from policy repository 212 and traffic repository 216, requested by controller 202. Accessing flow control system 200 and controller 202 using, for example, a graphical user interface (“GUI”), users can monitor, modify, delete, and enter user-configurable information at usage collector 204. Graphical representations that may be indicative of data traffic along one or more routes (e.g., paths between a source and a destination) can also be presented to users via a GUI. Network or systems administration personnel, or any entity responsible with flow control, with access to source network 82 can provide control information to flow control system 200 via usage collector 204 to modify system operation by, for example, changing data traffic flow from an under-performing route to a route which is more efficient route. Intervention by network administration personnel, however, is not necessary for flow control system 200 to operate in accordance with the present invention.
Flow control system 200 also compares specific data traffic flows (i.e., both uni- and bi-directional traffic flows outbound from and inbound into the data network) to determine whether a particular traffic flow meets one or more rules of an associated flow policy. A flow policy, as referred to herein and stored in policy repository 212 includes a set of one or more rules that is associated with a particular data traffic flow related to particular system user (e.g., as denoted by IP address prefix). Specified rules set by either default or user-configured are also stored in policy repository for providing parameters on the performance and operation of flow control system 200. Criteria for performance are used to guide the operation of flow control system 200.
Minimum criteria represent the lower levels of a range of values that defines acceptable routing behavior associated with a particular traffic flow characteristic. For example, a rule can set: the maximum acceptable cost, with or without regard to network service provider cost; the maximum load or bandwidth usage associated with traffic flows through specific providers; a range of acceptable (or non-acceptable) service providers; the maximum acceptable latency or loss over one or more paths across multiple network service providers; acceptable ranges of performance for each network service provider, such as maximum burst limits, minimum performance commitments and range of costs (i.e., cost structures with regards to time of day, type of traffic, etc.); and any other data flow characteristic that can influence the measurement or the control of data traffic.
Flow control system 200 further operates to detect when one or more rules, or flow policies, are violated and then to take remedial action. That is, flow control system 200 enforces policies associated with data traffic flow by correcting detrimental deviations in performance (i.e., service level assurance), costs or bandwidth (i.e., load in terms of percent capacity available per path). Flow control system 200 makes such corrections based on real- or near-real time traffic analysis, local path diversity (i.e., modifying one or more egress paths from a data network), and visibility into downstream available paths. For example, for a destination related to a specific traffic flow, flow control system 200 directs, or re-directs, traffic to one or more alternative paths to resolve a particular flow's deviation in terms of flow characteristics, from its flow policy.
In another embodiment, flow control system 200 can be a reactive flow control system. That is, a reactive flow control system is designed to react to policy violations indicating sub-standard routing of data traffic over one or more data networks or service providers (i.e., addresses pass-fail criteria) rather than optimizing performance at some targeted level of acceptable operation.
Referring back to
Apart from managing traffic flow, flow control system also uses data stored in traffic repository 216 and policy repository 212. Both traffic repository 216 and policy repository 212 are databases, and can be implemented using devices such as a storage device, database, storage enclosure, or other storage and database applications, data warehousing, and database management systems similar to those manufactured by companies such as Microsoft (MySQL), Oracle (9i+), Sybase, and EMC. Both traffic repository 216 and policy repository 212 are configured to store a large number of records in one or more data structures.
Traffic repository 216 is designed to store and to communicate information related to traffic and route characteristics, and policy repository 212 is designed to store and to communicate flow policy information or rules to govern the performance and cost of each of the data traffic flows. One having ordinary skill in the art of database management should appreciate that many database management techniques may be employed to effectuate the repositories of the present invention.
In operation, flow control system 200 of
Passive calibrator 214 monitors the data traffic of data flow 218 and communicates information related to the traffic, traffic performance, and policy adherence to controller 202. Controller 202 is configured to receive policy data which may represent one or more policies that correspond to a particular traffic flow, such as a particular data flow. Moreover, the particular data flow can be associated with a certain user identified by a destination prefix, for example. From policy data, controller 202 determines the levels of performance, cost, or utilization that the particular traffic is required to meet. For example, controller 202 determines whether a particular traffic flow of data flow 218 is meeting defined performance levels (i.e., service levels) as defined by one or more requirements or criteria, such as inbound and outbound network latency, packet loss, and network jitter.
Active calibrator 210 functions to send and to receive one or more active probes, of varying types, into and from the data networks. These probes are designed to measure network performance including, path taken across one or more available providers (i.e., to determine if a provider is a transit AS rather than peer AS), next hop-in-use, and other network parameters. To activate active calibrator 210, controller 202 sends an active probe request to active calibrator 210. Such a request is required if controller 202 determines that additional information regarding alternative routes or network system characteristics are necessary to better enforce policies in reactive flow control systems, or alternatively, to prevent such policy violations optimized flow control systems.
Usage collector 204 is configured to receive NSP data via data director 206 which communicates with one or more NSPs, depending upon the type of NSP configuration. Generally, such configurations include the number of route (“pipes”) associated with each provider and the data transfer rates, bandwidth allocations, redundancy provisions, data backup, and other performance characteristics associated with the NSP. Additionally, NSP data can relate to a provider's cost or billing structure and can also include each provider's associated set or subset of addresses, each provider's billing methods (i.e., byte/min, etc.), etc. Moreover, usage collector 204 is configured to collect usage information from the network elements, such as switches, border routers, provider gear, and other devices used to transport data over data networks. Usage collector 204 is configured to provide controller 202 with provider utilization and billing information, which represents aggregated data based upon NSP data and usage information. Often, utilization and billing information provide the basis for performance and operating parameters for flow control system 200. Usage data or utilization data (not limited to those embodiments described herein) can be measured in various ways to include connection or link time, data packet volume, or other metrics not included here specifically. Other utilization and billing information includes data that represents cost, billing, utilization, etc., for each NSP of interest.
One having ordinary skill in the art should appreciate that NSP data can be provided to usage collector 204 in a variety of ways. For example, data describing, defining, controlling, or modifying data paths as used for the data flows described herein can be provided by an entity having authority to do so, such as a network engineer entering the data into a computing device using a GUI in source network 82 of
Moreover, usage collector 204 is configured to monitor usage characteristics defining a network service provider's data traffic capacity, costs, etc. Usage data provided to usage collector 204 includes usage information and characteristics from network elements, such as switches, border routers, routers, provider gear, and other devices used to transport data over data networks. Usage refers to the data (i.e., data such as X Mb samples at time(0)) that represents instantaneous or near instantaneous measurement of characteristics (i.e., usage characteristics, data volume) that define, for example, the load and available capacity of each network service provider. Utilization is the usage rate over time. For example, suppose the usage collector monitoring NSP1 measures its utilization, or capacity over time, as X Mb at time(0) and Y Mb at time(1). This raw data, or usage, is used to calculate utilization, or usage rate for NSP1 (e.g., Y−X/time(1)−time(0)). Bandwidth is the total capacity each path or segment of path available for traffic flow. In one embodiment, the usage can be measured in any segment in any path at any number of hops or networks from a first point. Load is typically defined as the amount of capacity a particular path is used to carry data traffic and can be expressed as load/bandwidth.
Usage collector 204 is designed to generate utilization and billing information based upon usage information and NSP data sampled from data traffic flow 218. Since each of the providers has different cost and billing structures, as well as different methods of determining usage costs, usage collector 204 operates to aggregate usage information accordingly to provide controller 202 with utilization and billing information. Software applications or other means may be used to implement utilization and billing information for financial accounting purposes.
Usage collector 204 then provides the utilization billing information to controller 202 for each NSP of interest. One having ordinary skill in the art should appreciate that the usage collector can provide additional information based upon the provider usage information, to the controller, as needed to better effectuate route control.
Controller 202 collects information (i.e., aggregated performance and usage characteristics) from each of passive calibrator 214, active calibrator 210, usage collector 204, and optionally traffic repository 216. Based upon the information collected, controller 202 determines a course of action that best alleviates the policy violations in respect to the information represented by policy data that is conveyed to controller 202. Once the course of action is determined, controller 202 initiates and sends a network routing change request to configuration element 208, which directs data director 206 to alter data traffic flow through assigned NSPs. In a specific embodiment, controller 202 also provides data representing one or more alternate data paths that can be used to resolve a policy violation.
Configuration element 208 is designed to communicate routing changes in the network to data director 206. Once configuration element 208 sends one or more routing changes, data director 206 then moves data traffic flow 218 from a current path to another path (e.g., from NSP1 to NSP2 to NSPn or a first path of NSPI to a second path of NSPI). Data director 206 thus operates to distribute traffic to these destinations across multiple network service provider links based on, for example, the cost and performance measured across each link.
In operation, configuration element 208 communicates one or more routing changes with data director 206, for example, by using a routing protocol such as BGP. Configuration element 208 functions to dynamically control routing behavior by modifying the source address of the traffic passing through data director 206. The source address is modified in a way that improves application performance as well as cost requirements.
The following discussion is a more detailed description of each of the elements of flow control system 200. Active calibrator 210 provides active mechanisms within flow control system 200 for determining the nature of downstream or upstream paths. This information is typically not available in any conventional protocol used on data networks such as the Internet, and must be collected external to the normal processes of networking. As shown in
In one embodiment, active calibration probes are communicated to available network or Internet routes via available data routes from data director 206. The returning active calibration probes are received via data director 206. Active calibrator 210 then forwards probe information back to controller 202, which contains performance information including alternate available paths, which is stored in traffic repository 216. Controller 202 then determines how best to enforce the specifics of the policy associated with the subject traffic flow. Exemplary active calibrator 210 employs active calibration mechanisms to provide, for example, long term statistics.
In another embodiment of the present invention, active calibrator 210 resides in data director 206 within, or alternatively, integrated with controller 202. There are several proprietary implementations of commercially available routers suitable to practice the present invention. One example of suitable active probes is the RMON probe. Cisco systems use Service Assurance Agent (“SAA”) that is derived from the remote monitoring (“RMON”) probes to send out active probes. SAA allows routers to measure and report network-originated application round trip times (“RTT”). Although not every probe mentioned below is available in SAA for network calibration, one skilled in the art would appreciate how each of the following might be implemented to practice one or more embodiments of the present invention.
An exemplary active calibrator 210 can use ICMP (Internet Control Message Protocol) echo request or other ping-type probes, lightweight TCP-based probes, Sting probes, “pathchar” probes, lightweight probes using User Datagram Protocol (“UDP”) packets with a predefined TTL (time to live), traceroute probes, or other active probes that are suitable for use by active calibrator 210 in accordance with the present invention. As used herein, “weight,” whether heavy or lightweight, is a local measure of whether a particular data route is preferred over another. A weight is given to a data route on a particular router using a route map and intended for use only by that router. Thus, the lightweight probes referred to herein indicate that they are intended for a specific pre-specified or directed router, per a route map created in accordance with an embodiment of the present invention.
These probes are received back by active calibrator 210 of
Exemplary passive calibrator 214 of
In another embodiment of the present invention, passive calibrator 214 examines the traffic stream in both directions (i.e., ingress and egress) and classifies each of the traffic streams into flows. Traffic flows, are monitored within passive calibrator 214 according to the underlying protocol state (e.g., such as regarding TCP sessions) over time. For example, passive calibrator 214 classifies the traffic flow according to round trip latency, percentage of packets lost, and jitter for each of the traffic paths. Such traffic route information is used to characterize the “end-to-end” performance of the paths carrying the traffic flows, which includes flow rates, and is aggregated into a series of network prefixes.
As described above, passive calibrator 214 is coupled to store, fetch and update traffic and route information stored in traffic repository 216 (connection not shown). Exemplary traffic repository 216 is a database configured to store and to maintain data representing traffic and route information that is useful to the end user employing a flow control system, such as flow control system 200 of
Such feedback will consist, for example, of types of traffic being sent, source addresses, destination addresses, applications, traffic sent by ToS or DSCP (“DiffServ Code Point”) setting (which might be integrated into a differentiated billing system), and volume of traffic. These statistics are fed into traffic repository 216 where, for example, a reporting engine or some other analysis application can access to the statistics. The information stored in traffic repository 216 is data representing such traffic route characteristics arranged in any suitable data structure as would be appreciated by one skilled in the art.
In one embodiment, passive flow analyzer 330 performs passive analysis on the traffic to monitor current traffic flow characteristics so the controller can determine whether the monitored current traffic flow meets associated policy requirements. Export flow analyzer 331 performs passive analysis on exported flow records from a network device, such as from those devices (e.g., router) that advertise traffic type, source and destination addresses, and other information related to the traffic that it travels across service provider links. An example of such a network device is Cisco's Netflow™ product. In another embodiment, passive flow analyzer 330 operates in accordance to the passive flow analyzer described in the above-mentioned U.S. patent application of Ser. No. 09/833,219.
Content Flow Analyzer 332 performs passive analysis of specific elements of data content, such as web site content. Export flow analyzer 331 and content flow analyzer 332 determine a set of relevant prefixes or a prefix list 334 that is associated with a specific user's policy. Prefix list 334 is sent as data representing such prefixes to an active detection process in the controller. Prefix list 334 can be one or more lists or data structures configured to store data representing performance and usage characteristics and are designed to be receive a query, for example, by the controller. Once queried, the passive flow analyzer provides the one or more prefix lists, or portions thereof, to the controller for use in determining a policy violation, for determining which routes or path comply with the flow policy, which path is the optimum path for routing data, and the like. An exemplary prefix list that can be generated by export flow analyzer 331 and content flow analyzer 332, as well as passive flow analyzer 330.
Export flow analyzer 331 and content flow analyzer 332 also are configured to notify controller 305 when a previously unseen prefix has been added to the prefix list 334. New prefix notification signal 335 enables the control element 1005 to establish a new baseline performance for this prefix and to seed the routing table with a non-default path, or alternative path (i.e., non-BGP), if necessary. In one embodiment, export flow analyzer 331 and content flow analyzer 332 provide for monitoring of performance characteristics.
Content flow analyzer 332 is typically used when the main source of traffic flow 340 is web site or other content. Content source 341 can be configured such that special or premium content 342 that must be optimized can be identified by the flow control system by using, for example, an embedded URL 343. URL 343 redirects the client to a content server running on the content flow analyzer 332. Content flow analyzer 332 receives a request for the content element, which is generally an image file (e.g., 1×1 GIF) and is invisible or imperceptible in relation with the main original content, and responds to the client with the content element 344. Content flow analyzer 332 then stores or logs this transaction in log files and, by using the log files, content flow analyzer 332 is able to perform aggregation and assemble content prefix list 334. The list 334 is passed along to controller 205, for example, for active service level monitoring and policy enforcement.
The small redirect URL acts to generate an HTTP request 420 in response to the small element of content. Content flow analyzer 432 sees this request 420 and responds 422 to it with, for example, a lightweight HTTP server 453. This server is fast and lightweight, and does nothing other than respond with the image file. The lightweight web server 453 logs the IP address of the client requesting the web page, and sends one or more IP addresses to aggregator 454. Aggregator 454 aggregates, or collates, individual IP elements 424 into prefixes of varying granularity (e.g., /8 through /32) and also aggregates the frequency that each prefix is seen over an interval of time.
That is, aggregator 454 classifies prefixes according to frequency of occurrence and provides aggregated (i.e., grouped) prefixes 426 to prefix list generator 455. Prefix list generator 455 creates destination prefix list 428 according, for example, to a prefix's importance in relation to the overall operation of the system as defined by the aggregated or grouped prefixes 426. For example, each monitored traffic flow is examined to determine the performance characteristics associated with a destination prefix or address.
Aggregate prefixes 426 are generally classified in terms of flow frequency, and average or total flow volume. Prefix list generator 455 sends updates to current prefix list 428 to controller 205 of
Referring back to
Destination prefix list 528 is generated by prefix list generator 552 by, for example, ranking and organizing traffic flow characteristics related to prefixes in order of relative importance. List 528 contains data representing an aggregation of prefixes prefix list 528 and is organized in determines the relevance, as determined by the system or an entity to ensure policy enforcement. For example, one or more prefixes can be ordered in terms of flow frequency and average or total flow volume in relation together prefixes available in the overall system. Prefix list generator 552 sends updates to the current prefix list to controller 205 of
Passive flow analyzer 630 uses a copy of the traffic 602 via a passive network tap or spanned switch port, as shown in
Additionally, passive flow analyzer 630 is designed to be “application-aware” according how each of the particular traffic flows is classified. Traffic can be classified according to the classifier described in the above-mentioned U.S. patent application Ser. No. 09/833,219. Passive flow analyzer 630 can inspect the payload of each packet of traffic 602 to interpret the performance and operation of specific network applications, such as capture and interpretation of the Realtime Transport Control Protocol (“RTCP”) for voice over IP (“VoiP”), for example.
Parser 651 is coupled to receive captured raw packets and operates to deconstruct the packets and retrieve specific information about the packet from each in the traffic flow. Exemplary parser 651 extracts information from the IP and TCP headers. Such extracted information from the IP headers include source and destination IP addresses, DSCP information encoded in the ToS (i.e., “type of service”) bits, and the like. DSCP carries information about IP packet QoS requirements. Each DSCP defines the Per Hop Behavior of a traffic class. DiffServ has 64 code points so that it can define 64 different types of traffic classifications. TCP header information includes source and destination port numbers, sequence number, ACK number, the TCP flags (SYN, ACK, FIN etc.), the window size, and the like.
TCP elements parsed from the TCP headers are especially useful in determining whether a policy is being enforced, in terms of performance. An increasing amount of traffic, however, does not rely on TCP and instead uses UDP. UDP does not contain the necessary information to determine service levels according to conventional approaches.
To determine service levels to these destinations, the present invention might employ a statistically relevant amount of collateral TCP traffic going to the same prefix or a series of active probes to the same destinations, or have the analyzer parse deeper into the packet and understand the traffic at the application layer (e.g., layer 7). There are some protocols running on UDP that have very specific requirements that are different from most other data traffic on the network. These protocols are loosely classified as “real-time” protocols and include things like streaming media and Voice over IP. Packet loss and latency, below a certain level, are secondary concerns for real-time protocols.
Most importantly, however, is reducing the variance in inter-packet arrival times (i.e., network jitter). Many real time protocols and standards such as H.323 as defined by the International Telecommunications Union (ITU) report the observed jitter in back channel communication known as the RTCP (“Real-Time Transport Control Protocol”), which is used to distribute time-dependent media data via IP multicast with feedback. If passive flow analyzer 630 of
Correlator 652 operates to interpret and to group the packet elements (e.g., TCP and IP) from the packets to determine the current service level of the flow and then groups the packets into a specific traffic flow. Flows are reconstructed, or grouped, by matching source and destination IP addresses and port numbers, similar to the process of stateful monitoring of firewalls. Correlator 252 determines the current service level by measuring several traffic characteristics during a TCP transaction. For example, correlator 252 determines the round trip time (“RTT”) incurred on a network, and hence, this serves as a measure of latency for the network traffic.
Usage collector 715 also operates to reconstruct provider billing records. Usage collector 715 accepts provider configuration information 771 related to each network service provider (NSP) connection. This NSP configuration information 771 details provider interfaces on the various routers 772 (e.g., egress routers), provider next-hop IP addresses traceroute probes (to verify the current provider in use with trace probes), billing period start and end dates, circuit bandwidth for calculating the utilization and price per megabit/sec, minimum bandwidth commitment, burstable rates, provider sampling interval, provider billing algorithm, a utilization alarm threshold and the like.
In operation, exemplary raw collector 774 sends a query 790 (e.g., SNMP) to collect interface raw byte counters from routers 772 on each of the provider circuits at a specified sampling interval. Provider circuits include paths, pipes (virtual or physical), T1, and the like. Raw Collector 774 places the raw byte counters 780 into persistent storage for later reporting and analysis. Raw collector 774 sends the raw information to two other components: utilization monitor 775 and bill reconstructor 776.
Utilization monitor 775 calculates the ingress and egress circuit utilization for each provider using the raw byte counts and the NSP configuration information 771. In one example, NSP configuration information 771 includes the bandwidth of the provider's circuits. Utilization information 764 includes data representing utilization trends for use with short range forecasting models (e.g., ARIMA, exponential smoothing, etc.) such that utilization monitor 775 can determine whether bandwidth is trending up or down (i.e., increasing or decreasing in size) for a given service provider.
Bill reconstructor 776 uses the billing information from NSP configuration data 771 to reconstruct the current provider billable rate for the current billing period. Billing information includes information explaining the methods that specific providers use to calculate costs, such as a billing rate. Such methods of calculating bills for using a network provider are well known in the art. Bill reconstructor 776 applies similar provider billing methods to the raw byte counters from raw collector 774 to generate the bill and related billing rates, etc. The generated bills, which are mapped into dollar amounts, are typically estimates since the sample times between the provider and usage collector 715 will not match exactly. Bill reconstructor 776 will send billing information 761 to controller 702 for use in peak avoidance and least cost routing. Peak avoidance is defined as a method of avoiding using a path or path segment at a higher a billing rate. Least cost routing refers to a method of using or defaulting traffic to the least expensive provider.
Additionally, information can be sent to controller 702 for use in the least cost fix method of selecting the cheapest if performance is of no consequence. That is, controller 702 uses data from billing message 761, including billing rates, to determine an alternate path based in part on a path's free bandwidth (i.e., path does not incur additional cost to use), in accordance with the flow policy.
In one embodiment, the routing tables are received into route server 791 by way of one or more default BGP feeds 792 into BGP Engine 782 from a full set or subset of the local transit providers. In
Referring back to
Configuration element 784 is designed to receive one or more BGP4 routing tables 783 from BGP4 engine 782 and is adapted to receive one or more control signals and data resulting from the control processes of controller 705. In operations, configuration element 784 receives, from controller 705, the necessary routing changes to be implemented in default routing table 783. Then, configuration element 784 incorporates one or more changes in modified routing table 789.
Thus, configuration element 784 operates to modify BGP4 routing table 783 and to generate one or more modified BGP4 routing tables 788. Modified BGP4 routing table 788 includes changed routing 789, advertisements of more specific routes, etc. New modified BGP4 routing table 788 is then fed to all BGP clients in the network, which then is used to guide traffic to the destination.
Each of the embodiments described in
Once data director 206 receives this information, controller 202 is informed that a route change has been implemented. Thereafter, controller 202 communicates with passive calibrator 214 to clear its state and to resume monitoring the destination. The destination is monitored to ensure that the updated path of the routing table, or path, meets minimum service levels (e.g., no violations of SLA, no unacceptable deviations from agreed upon performance metrics as defined by policy, criteria for best performance, or other flow policy exceptions, as discussed below with regard to events in controller 202 logic).
In one aspect, configuration element 208 enables configurations to data traffic flow 218. Configuration element 208 may reside within a server, security tunnel, network switch, IP security device or application, load balancer, firewall, or other security-enabled device for use with a network used when connecting one or more paths between a destination and a source or sending modified traffic through a single path. In another aspect, configuration element 208 resides in a router and is configured to modify a route map or table. In yet another aspect, configuration element 208 is adapted to provide configuration information which may be stored in a data structure such as a routing table. In still yet another aspect, the path information is stored within the configuration element 208 according to whether it is related to inbound or outbound traffic.
For a given source address, the ingress point into a network is determined typically by the advertisements of routes made to downstream providers and a provider policy (set of rules that is set up by such providers). Eventually, the network service provider (e.g., “NSP,” “ISP”) that is hosting the destination will receive advertisements, announcing preferred routes, destinations, paths, and routing changes made by controller 202 in accordance with an embodiment of the present invention. For example, in one embodiment of the present invention, advertising or announcing a route prefix of “/16” would indicate that only the first 16 bits of the address are relevant and, given that 32 bits are available to describe a specific address, that all related and subsequent addresses to that prefix would also be considered as part of the advertised path.
Controller 202 of
If a particular policy is violated (i.e., one or more performance and/or cost metrics are outside one or more expected ranges or values), controller 202 determines a subset of one or more alternate data paths that conform to policy. In another example, controller 202 selects a best or optimized path as an alternate data path that best meets the performance requirements and usage requirements, (e.g., billing profiles and contracts, cost policies per individual NSP, static and dynamic policies either system-specified or user-configured) as defined by the policy. The following discussion relates to policies and the application of such policies in resolving policy violations, enforcing the policy requirements or metrics, and policy examples.
Referring back to
Detection is defined as the techniques or mechanisms by which flow control system 200 determines which traffic that should be acted upon in response to policy. The traffic flow can be identified, by name, by source or destination addresses, by source or destination ports, or any other known identification techniques. For example, a policy can be identified by address prefix. That is, flow control system 200 will monitor the traffic flow to and from a specific prefix, and if necessary, will enforce policy in accordance to its requirements. Further regarding detection, a policy defined for more specific prefixes can take precedence over more general prefixes. For example, a policy defined for a /24 will take precedence over a /16 even if the /16 contains the specific /24.
Performance is a policy requirement that describes one or more target performance levels (i.e., network/QoS policy parameters) or thresholds applied to a given prefix, prefix list, or other identified set of destinations (e.g., identified by application, port, protocol, etc.). Although more than one performance-based policy requirement may be defined, in this example only a single policy is applied to a given prefix or prefix list. Exemplary performance requirements include loss, latency, and jitter.
Moreover, such requirements can be configured either as, for example, an absolute, fixed value or as an Exponentially Weighted Moving Average (“EWMA”). In another embodiment, absolute value establishes a numerical threshold, such as expressed as a percentage or in time units over a configurable time window. The EWMA method establishes a moving threshold based on historic sampling that places an exponential weighting on the most recent samples, thereby asserting a threshold that can take into account current network conditions as they relate to historic conditions.
In another embodiment, requirements can be configured using Statistical Process Control (“SPC”) techniques. SPC techniques are leveraged to determine when the network performance for a given prefix deviates from the observed and desired performance. SPC generates upper and lower control limits (UCL, LCL) based on the historical mean performance. The UCL/LCL is a fixed delta (e.g., ms) from the long running observed mean. However, RTT is tightly correlated to geography, and this prohibits the simple applications of SCP where UCL/LCL is determined using a ‘fixed’ delta (e.g., 10 ms) or a fixed percentage of the observed mean. (e.g., 10% deviation is considered an exception).
Most process control techniques assume data that is normally distributed, in which case there are a myriad of process control charts that can be applied. However, performance metrics such as network latencies are not normally distributed and as such the standard process control charts, which define deltaU and deltaL cannot be used. DeltaU and deltaL are used from the mean to calculate UCL/LCL. As such the process control chart which defines UCL/LCL is unique for different performance values. This results in a banded approach for determining a unique process control chart and hence, UCL/LCL. Volume profiles can be defined around bands or values and will be discussed in greater below with regard to
In the purest sense, the function that determines deltaU/deltaL is a non-linear function of the mean RTT. In yet another embodiment of the process control chart, this function was approximated for several mean performance values such as RTT space. This results in a single continuous process control chart that is relevant to all performance values.
UCL and LCL define the operating range where two sets of performance numbers should be considered ‘equivalent.’ It is only when a candidate network performance is outside the operating region defined by UCL/LCL that a change is warranted. This will prevent a situation where a slight difference in a performance metric, such as RTT or loss, can cause a route change.
Network performance ‘Exceptions’ occur when several performance samples are observed above the UCL or below the LCL. Differences between the observed samples and the UCL and LCL which do not exceed the UCL/LCL are considered to be within a range of equivalence. However, exceptions, when defined as a number of samples above or below the UCL/LCL, indicate that the underlying system (i.e., the network) has changed in a fundamental manner, (e.g., system performance improves or degrades such that UCL/LCL are exceeded). For example, in the event of an Exception, downstream flow control policy components can examine candidate paths and determine if better performance is available from an alternative provider.
While the configuration of the SPC varies, an example default configuration for exception detection of an RTT performance metric is:
Cost can be expressed in a policy as a requirement or rule in terms of precedence and whether the policy is predictive or reactive. Costs are characterized by usage collector 214 of
Typically, both cost predictive and reactive requirements result in a binary decision (i.e., a circuit or path, for example, is either in compliance with or in violation of a flow policy). In the case of predictive cost, the transit circuit is either in compliance, or soon to be violation of a flow policy. Regardless, an action must be taken to resolve the situation, unless cost is preceded by performance (i.e., performance requirements are to be addressed prior to making a cost-based change).
Precedence is a policy requirement that describes one or more target usage or utilization characteristics or levels. Precedence includes provider preference and maximum utilization (i.e., load) requirements. The provider preference requirement is, for example, an arbitrary ranking of providers that is used when an action must be taken, but when two or more transits may be selected in order to enforce the policy. The flow control system can automatically set the provider or path preference requirement if it is not configured explicitly by the system's operator. This requirement is then applied as a tiebreaker in deadlocked situations such that the provider with the highest preference is selected to receive the diverted traffic flow.
The maximum allowable utilization can be used as an actual operational threshold not to be exceeded or as a tiebreaker. Maximum usage is configured, for example, in the transit provider section of the configuration and takes either a percentage argument (i.e., in terms of available bandwidth), or alternatively, can be set as an absolute value in terms of Mb/s (i.e., not to exceed available bandwidth).
Policy examples, in addition to those described above, can be applied to each destination and may include minimum performance metrics (e.g., loss, latency, jitter), and measurement behavior (e.g., duration, exclusion, frequency, RTT, etc.). Other policy examples include the ability to specify routing changes on/off (i.e., specified parameters enabling an embodiment of the present invention to record results, but make/not make routing changes, if criteria are met), decision criteria (e.g., cost/performance priority), thresholds (e.g., dynamic, static) or other criteria which can be configured and stored in policy repository 212. In accordance with an embodiment of the present invention, dynamic thresholds are based upon a historical baseline, from which all derived information goes into a repository or data structure. Static thresholds, on the other hand, are arbitrary user-defined baselines which provide the basis for establishing, for example, a static performance threshold with minimum performance levels.
In summary, the policy, such as the above exemplary policy, is input into the controller 202 of
The active calibrator and the passive calibrator provide performance characteristics. Regarding the active calibrator, controller 202 initiates active calibration by requesting active probing. The active calibrator sends one or more calibration probes out into the one or more data networks. Responsive information received in response to the probes provides information back to controller 202, which contains the identities of available paths and performance information related thereto.
Regarding passive calibrator 214, controller 202 is designed to receive real- or near-real time network performance characteristics (i.e., loss, latency, jitter, etc.) from passive calibrator 214 as monitor in traffic flows in which it has access. In accordance with an embodiment of the present invention, controller 202 can be set to only record changes. However, in another embodiment of the present invention, controller 202 can be set to record and make changes to traffic flows, based upon real-time network performance characteristics. Control signals initiate clearing the passive calibrator 214 state to enable the passive calibrator 214 to resume monitoring of the specific destination to ensure that the updated route of the routing table, or path, is complying with recorded flow policies. Control signals to clear the passive calibrator 214 state come from the controller 202 to initiate the resetting of the passive flow analyzer 330 state. However, in other embodiments of the present invention, control signals may not be necessary to clear the passive calibrator 214.
In an example, controller 202 operates to interpret candidate performance for alternative performance for newly observed or aged destination prefixes. Controller 202 determines which of the alternate path or paths, is best suited for the prefix or traffic type associated with the current traffic flow. As discussed above, controller 202 can also interpret candidate performance in newly observed or aged destination prefixes, and make changes to traffic flows accordingly. Controller 202 then sends the necessary routing changes to configuration element 208. That is, controller 202 finds the best initial baseline in accordance with policy. This process is repeated if the prefix ages again without subsequent measurement. If the prefix is also listed as either an exception, a high volume prefix, or a very important prefix, controller 202 can also make necessary routing changes to configuration element 208.
In another embodiment, controller 202 operates to interpret candidate performance for alternative performance for high volume destination prefixes. These prefixes are significant when compared against lower data volume prefixes. Controller 202 determines which of the alternate path or paths, is best suited for the prefix. Controller 202 then sends the necessary routing changes to configuration element 208. That is, controller 202 finds the best performance in accordance with policy. This process is repeated until the prefix fails to meet high volume criteria stored in the policy repository, e.g., volume levels (such as top “n” level, i.e., top 100%), flow levels, or other system configurations specified and stored in the policy repository.
In yet another embodiment, controller 202 operates to interpret alternative path performance for destination prefixes that have static persistent prefix(es). These prefixes are significant as specified by the user and are persistently measured. Controller 202 determines which of the alternate path or paths, is best suited for the prefix associated policy. Controller 202 then sends the necessary routing changes to configuration element 208. That is, controller 202 again finds the best performance in accordance with policy. This process is repeated as long as the static persistent prefix exists.
In a further embodiment, controller 202, as depicted in
To resolve usage-type policy violations, controller 702 is configured to receive routing tables, for example, to determine which of the current traffic flows or routing of data on certain paths, or path segments thereof, are congested (i.e., loaded) with respect to a particular provider path or paths. Controller 702 receives flow data to determine which flows are active on the NSP which is indicated as “full” (e.g., no available bandwidth). Controller 702 also uses flow data to determine the volume of the active flows, aggregated into prefix volumes. The controller 702 also uses the flow data to determine the stability of the flow volume. Stability is, in accordance with an embodiment of the present invention, desirable, particularly when a cost change is made, but remains unaffected when the traffic flow volume decreases or ceases. From the flow data, controller 702 can determine which prefix(es) and the number of prefix(es) that need to be moved in order to correct an exception, e.g., a cost violation. Also important is flow volume projection.
Projecting flow volume is important when making changes in a cost-constrained environment. If a flow is being moved because of cost constraints, understanding the volume to and from a given NSP affects the accuracy of the decision. In an embodiment of the invention, it is possible to scale the observed volume on the existing provider based on the current and projected RTTs to estimate flow volume at the destination. The delay bandwidth product governs that flow volume for TCP is a function of window size and RTT. If average window size remains constant across providers, flow volume can be scaled by using RTT. For example:
Thus, controller 702 is designed to obtain information related to the performance and usage of data networks and to implement corrective action to achieve the NSP policy (e.g., lowest network cost, load balancing, etc.) in data routing over flow control system 200 in data traffic flow 218 via associated NSPs.
In order for controller 202 to handle peering connections, controller 202 communicates with data director 206 to retrieve reachability information (i.e., routing tables) for the specific prefix that is about to be changed. In the case of transit circuits, controller 202 uses active calibrator 210 to determine reachability information (i.e., routing tables) for a given destination by, for example, sending active probes to the destination and then waiting for the response. Although peering connections are often unreachable, it is possible for active probes to succeed since some providers may not effectively filter traffic at a peering point and instead rely on an honor-like system to ensure that only traffic to those advertised destinations is received.
Therefore, in the case of peering or private connections, controller 202 must look in the routing table for an advertisement of that destination before moving traffic to a peering connection. Alternatively, a set of reachable destinations can be statically configured for a provider/NSP peer or private link (NSP). Controller 202 can look in the static configuration for reachability information, stored as part of the system configuration. This allows support for public and private connection where routing information may not be dynamically available over private connections.
In an embodiment of the present invention, cost and/or usage control and best performance goals are sought. For example, given a diverse set of cost structures for available transit providers, the overall flow control system cost can be minimized by distributing traffic in a way that takes advantages of the diverse billing structures and yet maintains acceptable performance levels. In another embodiment of the present invention, balancing load or usage across multiple NSPs may be desirable. A model of overall system cost is the minimum cost curve shown below representing peak usage levels for, as an example, three NSPs plotted against the total costs for all three NSPs, A, B, and C in
Referring back to
The three providers (NSP A, B, and C) have three simple billing profiles. Each NSP has a minimum commitment followed by a burstable tier. The cost/data transfer rate (i.e., Mbps, Gbps) is distinct for each NSP. NSP A has the lowest cost/data transfer rate in the burstable tier, NSP B has the next lowest, and NSP C has the most expensive cost/data transfer rate in the burstable tier.
With regard to cost enforcement in accordance with an embodiment of the present invention, the maximum usage levels on each provider can be determined from the minimum cost curve as shown in
In the graph as shown, the minimum commitment level should be enforced for NSP C, the maximum utilization should be enforced on NSP A, while an intermediate level between the minimum commitment and the maximum utilization should be enforced on NSP B. From the graph, a representation of the level that should be enforced per each NSP as 3 volume profiles as shown in
The volume profiles shown below in
The least cost curve is derived from the billing profiles of the different egress providers. As an example, each egress has 2 price tiers, the first of which is a fixed cost, fixed bandwidth or minimum commitment tier. The second price tier has a marginal cost for additional P95 bandwidth. For the flow control system relevant to the embodiment shown in
If precise usage samples are known for a particular billing period, P95 bandwidth for the sum of NSP bandwidths can be derived. This derived P95 total NSP bandwidth (total NSP bandwidth is used to distinguish individual NSP bandwidth from the sum of bandwidth across all egresses) as the bandwidth that determine how to distribute traffic across different NSPs. Referring to FIG. 8C's expected P95 line, it is apparent that it crosses the least cost curve in cost tier B2. Thus, the ideal allocation of bandwidth for cost reasons would be:
Available for Use
100% Tier A1 +
100% Tier A2
100% Tier B1 +
70% Tier B2
100% Tier C1 +
0% Tier C2
If the P95 total NSP bandwidth is below the sum of the output bandwidths of the minimum commitment tiers (A1+B1+C1), then the table would appear as:
Available for Use
100% Tier A1 +
0% Tier A2
100% Tier B1 +
0% Tier B2
100% Tier C1 +
0% Tier C2
If the current total bandwidth is above the expected P95 total bandwidth, then bandwidth allocations are computed along the Least Cost Curve provided current total bandwidth remains above the expected P95 point. The cost usage model/algorithm then addresses the decision, if costs are optimized, whether, in light of marginal ability for performance optimization, the controller should swap prefixes between NSPs. In conventional data route control techniques, particularly where BGP is used, a “worst case” scenario tends to evolve. In other words, BGP tends to distribute data traffic flow so as to follow the shortest path, and not necessarily the best performing/lowest cost path available. Often, the best path selected by BGP is an overly-expensive tier one NSP.
Unlike the above-discussed embodiment of the present invention, BGP can select the most expensive path, simply because it is the shortest path, and thus significantly reduce the efficiency of data routing for a multi-homed organization with regard to cost. Disregarding any analysis of P95 bandwidth allocations, neither cost nor performance information is taken into account by BGP when allocating bandwidth among numerous providers. Additionally, BGP also fails to adjust data traffic allocations, even when a significant degradation to either cost or performance appears. Ultimately, an exemplary cost curve for BGP-based route control follows the maximum cost curve in
As for computing bandwidth (as indicated at 7 on
The above discussion and embodiment relates to outgoing bandwidth allocations. However, this discussion is not intended to limit the invention and other embodiments of the present invention can address input traffic bandwidth allocations. Where NSPs bill based on input bandwidth, output bandwidth, max (input bw, output bw), or sum (input bw, output bw), the bandwidth allocation scheme can be improved with a richer filling method. The richer filling method is accomplished by using P95 bandwidth (or P other estimate, e.g., 87%, 80%, etc.) estimate of the input bandwidth to fill the egresses first. The egresses are filled using the P95 (or other percentile) input bandwidth estimate first because there is no immediate control over the input path. The rules governing the richer filling method, as an example for in-bound traffic, are:
As an example of an outbound traffic richer filling method, the above-rules would be conversely modified, for example, as:
Richer filling is required when configuration element 208 (e.g., BGPd) is only able to optimize either inbound or outbound usage. For example, in accordance with one embodiment of the present invention, BGP can only affect outbound usage per destination. This requires richer filling techniques for inbound when inbound is considered in the billing profile (e.g., max (in, out)). Some embodiments of SPRE or other NAI-based configuration elements can only affect in-bound usage per destination. This requires richer filling for outbound when outbound usage is considered in the billing profile (e.g., sum (in, out)). If configuration element 208 can affect both inbound and outbound usage, then no rich filling technique is required.
Usage data, upon request, can be communicated to control module 904. Flow cache 924 stores current path ID from a lookup in RT Mirror 928 for destinations. Flow performance information is passed to process control module (PCL) 925, which is designed to determine performance violations using SPC, EWMA, static or other methods. If a violation is detected for a destination, a violation flag is set in flow cache 924 and sent to controller 904. With regard to passive calibrator 910, PFA data can be communicated to control module 904 and, via flow cache 924. PFA 914 data is generated on a per-destination basis, e.g., Round Trip Times (RTT), loss, volume, and jitter. Route table mirror 928 is also communicating with route cache 930, which is sending data from the configuration element 912. Data gathered by configuration element 912 can include, but is not limited to, updates, withdraws, and other routing information from external BGP peers, other protocol-based information, and self-generated information, in accordance with one embodiment of the present invention. Although shown as a BGP-enabled device/process in the current embodiment, other embodiments may use protocols known to those skilled in the art and the present invention is not intended to be limited by only the embodiment currently illustrated.
Referring back to
Information gathered from active calibrator 914 may include, but is not limited to, convergence point analysis information, per prefix/destination information (e.g., RTT, packet/data loss, jitter, error codes, egress IDs, etc.), or any other type of information that embodiments of the present invention may gather. This information may also be communicated to control module 904 in order to enable route control and changes as necessary. Policy repository 942 houses policy, rules, parameters, and other user or system-specified criteria under which controller 902 provides route control. Control module 904 may also use consolidated usage information plus egress cost information from usage module 918 and egress cost module 920, respectively. In doing so, control module 904 is able to determine whether cost violations exist, based upon policy input received from policy repository 942 which, again, may be either user or system-specified.
Further, in accordance with other embodiments, the present invention is not necessarily limited to those modules shown in
On NSP A 1002, spheres 1008 and 1010 represent performance and cost-constrained bandwidth allocations according to policy priority settings, respectively. Spheres 1008 and 1010, if moved, may incur either additional costs or performance effects as a result of moving between NSPs. However, cost and performance priority policy setting kicks in only when there is a constraint. Performance priority is not allowed to be degraded in order to address cost moves. Instead the movement of other flows will be selected. Likewise, cost priority is not allowed to increase cost to improve performance. In NSP A 1002, a minimum commitment 1004 is shown, with no bandwidth allocation spheres exceeding the minimum commitment 1004. Also shown are a usage threshold 1006 and a maximum allowable utilization which represents the difference between the maximum bandwidth and the usage threshold 1006. For example, with a 100 Mbps provider and maximum utilization of 90%, the maximum allowable utilization would be 100-90 or 10 Mbps. NSP B also shows a minimum commitment 1012, with no bandwidth allocation spheres exceeding the minimum commitment 1012. Also shown are a usage threshold 1014 and a maximum allowable utilization 1016. Also shown in NSP B are cost-constrained bandwidth allocation spheres 1010. Finally, NSP C also shows a minimum commitment 1020, with no bandwidth allocation spheres exceeding the minimum commitment 1020. Although not represented, NSP C 1006 does have a usage threshold, as determined by the model in
Spherical bandwidth allocations, some of which are shaded, represent in-use bandwidth for various applications to a destination prefix. In particular, the shading of bandwidth allocation sphere 1008 is performance-constrained and bandwidth allocation sphere 1010 is cost-constrained by policy. The shading of the spheres also represents the relative performance to a historical perspective or baseline. When performance deviates from the perspective an exception, as discussed above, is indicated, representing bandwidth allocations that should be moved due to performance constraints. The constraints are established and placed on each of the bandwidth allocations with regard to policy input and policies stored in the policy repository 942 (
In terms of cost constraints,
In another example illustrated in
As shown in
In step 1302, controller 202 evaluates current latency for a destination address or prefix between the evaluated candidate NSP (e.g., egress router) and a destination address or prefix. If controller 202 determines that current latency is less than latency available via the candidate NSP, then no change signal is sent to the evaluated NSP. However, if current data transmission latency at egress exceeds latency available via candidate NSP, controller 202 then evaluates candidate and current loss in step 1306. If candidate loss is less than current loss, then controller 202 determines whether candidate's current bandwidth allocation is less than the total bandwidth allocation. If so, then the candidate evaluation continues, as discussed with regard to
If, in step 1308, the current candidate bandwidth allocation is less than the total bandwidth allocation for the candidate (according to the cost-usage model), then controller 202 evaluates the candidate NSP against all other candidate NSPs. In
If the candidate NSP does not have the greatest available bandwidth, then no change signal is sent to the evaluated NSP. If the candidate NPS does have the greatest available bandwidth, then a change signal is sent to the evaluated NSP. If the available bandwidth at the evaluated NSP is equal to another candidate NSP, then controller 202 must determine the next least cost tier in step 1324. In this step, controller 202 must determine which of the equal NSPs resulting from step 1322 has the lowest next cost tier and subsequently, direct a change to that NSP. If the resulting candidate NSPs from step 1322 is in the same next cost tier, then, in step 1326, the NSP with the lowest egress ID will be chosen and controller 202 will send a change signal directing data flow to the candidate NSP.
Similarly for performance criteria, controller 202 logic will execute the flow chart as shown in
In step 1342, if any NSP has available bandwidth, then controller 202 determines, in step 1344 whether the candidate is the least cost link. If the candidate NSP is the least cost NSP, then a change signal is sent to the evaluated NSP. If not, then no change signal is sent to the evaluated NSP. If the cost is equal between the candidate NSP and other candidate NSPs, then controller 202 assesses whether the candidate NSP is the lowest egress ID, as shown in step 1346. If not, then no change signal is sent. If so, then a change signal is sent.
Referring back to step 1342, if no candidate NSP has bandwidth available, then controller determines whether the candidate NSP has the most bandwidth available in step 1348. If so, then a change signal is sent to the evaluated NSP. If not, then no change signal is sent. However, if the available bandwidth of the candidate NSP is equal to another candidate NSP, then controller determines whether the candidate falls within the next cheapest cost tier. If so, a move is made to the candidate NSP. If the cost of the candidate NSP is equal to other candidate NSPs, then again, the lowest egress ID assigned will be selected as the best performing NSP. The above-process outlines the method by which, in accordance with one embodiment of the invention, the controller 202 may execute the logic behind best performance path selection. One of ordinary skill in the art, however, may be able to determine alternative steps, processes, sub-processes, or methods by which best performance path selection can be made and the present invention is not intended to be limited to necessarily those steps stated above.
As described, embodiments of the present may send, generate, use, exploit, gather, intercept, receive, manipulate, modify, destroy, or propagate signals to transfer data. The data network 82 (
As is well known in the art of network communications, data networks are configured to communicate electrical information, such as a computer data signal comprising data (e.g., binary data bits) superimposed upon a radio or any other carrier wave. A person having ordinary skill in the art would appreciate that a carrier wave is electromagnetic energy propagated from a source by radiation, optical or conduction waves and is suitable for embodying an information-bearing signal, such as a computer data signal. In one embodiment, a carrier wave behaves, or is modulated, according to a network protocol, such as or Ethernet, IEEE 1394, TCP/IP, or any other communication protocol, so as to include computer data information. In some embodiments, the computer data signal includes a computer program for practicing the present invention. The carrier wave can be, for example, a direct current, an alternating current, or a pulse chain. In modulation of the carrier wave, it may be processed in such a way that its amplitude, frequency, or some other property varies so as to embody data for transfer.
Although the present invention has been discussed with respect to specific embodiments, one of ordinary skill in the art will realize that these embodiments are merely illustrative, and not restrictive, of the invention. For example, although the above description describes route control as relating to Internet data, it should be understood that the present invention relates to networks and route control in general and need not be restricted to Internet data and signals, as discussed above. The scope of the invention is to be determined solely by the appended claims.
In the foregoing specification, the invention is described with reference to specific embodiments thereof, but those skilled in the art will recognize that while the invention is not limited thereto. Various features and aspects of the above-described invention may be used individually or jointly. Further, although the invention has been described in the context of its implementation in a particular environment and for particular applications, its usefulness is not limited thereto and it can be utilized in any number of environments and applications without departing from the broader spirit and scope thereof. The specification and drawings are, accordingly, to be regarded as illustrative rather than restrictive.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5598532 *||Oct 21, 1993||Jan 28, 1997||Optimal Networks||Method and apparatus for optimizing computer networks|
|US5724513||Jun 30, 1994||Mar 3, 1998||Digital Equipment Corporation||Traffic shaping system for asynchronous transfer mode networks|
|US5781534||Oct 31, 1995||Jul 14, 1998||Novell, Inc.||Method and apparatus for determining characteristics of a path|
|US5870561||Mar 15, 1996||Feb 9, 1999||Novell, Inc.||Network traffic manager server for providing policy-based recommendations to clients|
|US5898668||Dec 13, 1996||Apr 27, 1999||Siemens Information And Communication Networks, Inc.||Method and system for increasing quality of service at or below a threshold cost|
|US5933425||Dec 3, 1996||Aug 3, 1999||Nec Corporation||Source routing for connection-oriented network with repeated call attempts for satisfying user-specified QOS parameters|
|US5953312||Sep 13, 1996||Sep 14, 1999||Bay Networks||Method and apparatus for determining alternate routes in a network using a connection-oriented protocol|
|US5974237 *||Dec 18, 1996||Oct 26, 1999||Northern Telecom Limited||Communications network monitoring|
|US6047326||Mar 20, 1997||Apr 4, 2000||Nokia Telecommunications, Oy||Accounting system and method for a nominal bit rate network service|
|US6047331||Feb 19, 1997||Apr 4, 2000||Massachusetts Institute Of Technology||Method and apparatus for automatic protection switching|
|US6055571||Nov 20, 1997||Apr 25, 2000||Nec Usa, Inc.||Computer network with microeconomic flow control|
|US6064677||Jun 10, 1997||May 16, 2000||Xerox Corporation||Multiple rate sensitive priority queues for reducing relative data transport unit delay variations in time multiplexed outputs from output queued routing mechanisms|
|US6078957 *||Nov 20, 1998||Jun 20, 2000||Network Alchemy, Inc.||Method and apparatus for a TCP/IP load balancing and failover process in an internet protocol (IP) network clustering system|
|US6097699||Jun 5, 1998||Aug 1, 2000||Gte Laboratories Incorporated||Method and system for monitoring broadband quality of services|
|US6130890||Sep 11, 1998||Oct 10, 2000||Digital Island, Inc.||Method and system for optimizing routing of data packets|
|US6154778||May 19, 1998||Nov 28, 2000||Hewlett-Packard Company||Utility-based multi-category quality-of-service negotiation in distributed systems|
|US6181679||Mar 19, 1993||Jan 30, 2001||International Business Machines Corporation||Management of packet transmission networks|
|US6185635||May 30, 1998||Feb 6, 2001||Alcatel Networks Corporation||Method and circuit for transporting data based on the content of ingress data words and egress data words|
|US6226751||Apr 17, 1998||May 1, 2001||Vpnet Technologies, Inc.||Method and apparatus for configuring a virtual private network|
|US6249801||Jul 15, 1998||Jun 19, 2001||Radware Ltd.||Load balancing|
|US6252848||Mar 22, 1999||Jun 26, 2001||Pluris, Inc.||System performance in a data network through queue management based on ingress rate monitoring|
|US6256314||Aug 11, 1998||Jul 3, 2001||Avaya Technology Corp.||Apparatus and methods for routerless layer 3 forwarding in a network|
|US6262976||Sep 17, 1999||Jul 17, 2001||Ordered Networks, Inc.||System and method for network flow optimization using traffic classes|
|US6275470||Jun 18, 1999||Aug 14, 2001||Digital Island, Inc.||On-demand overlay routing for computer-based communication networks|
|US6292464||Sep 17, 1997||Sep 18, 2001||Nortel Networks Limited||Apparatus and method for self routing control mechanism for restoring fiber optic communications network connections|
|US6363053 *||Feb 8, 1999||Mar 26, 2002||3Com Corporation||Method and apparatus for measurement-based conformance testing of service level agreements in networks|
|US6363056||Jul 15, 1998||Mar 26, 2002||International Business Machines Corporation||Low overhead continuous monitoring of network performance|
|US6374300||Jul 15, 1999||Apr 16, 2002||F5 Networks, Inc.||Method and system for storing load balancing information with an HTTP cookie|
|US6393486||Oct 28, 1999||May 21, 2002||Cisco Technology, Inc.||System and method using level three protocol information for network centric problem analysis and topology construction of actual or planned routed network|
|US6400681||Apr 1, 1997||Jun 4, 2002||Cisco Technology, Inc.||Method and system for minimizing the connection set up time in high speed packet switching networks|
|US6404769||May 18, 1998||Jun 11, 2002||Motorola, Inc.||Directional routing of packets in a satellite network|
|US6405219||Sep 24, 1999||Jun 11, 2002||F5 Networks, Inc.||Method and system for automatically updating the version of a set of files stored on content servers|
|US6466984||Jul 2, 1999||Oct 15, 2002||Cisco Technology, Inc.||Method and apparatus for policy-based management of quality of service treatments of network data traffic flows by integrating policies with application programs|
|US6496477||Apr 19, 2000||Dec 17, 2002||Texas Instruments Incorporated||Processes, articles, and packets for network path diversity in media over packet applications|
|US6549781 *||Apr 12, 2000||Apr 15, 2003||Verizon Laboratories Inc.||System and methods for rapidly determining CDMA network performance|
|US6574195||Apr 19, 2000||Jun 3, 2003||Caspian Networks, Inc.||Micro-flow management|
|US6625150 *||Dec 16, 1999||Sep 23, 2003||Watchguard Technologies, Inc.||Policy engine architecture|
|US6658457||Nov 18, 1999||Dec 2, 2003||Fujitsu Limited||Device and method for interconnecting distant networks through dynamically allocated bandwidth|
|US6665702||Dec 20, 1999||Dec 16, 2003||Radware Ltd.||Load balancing|
|US6678248||Jun 20, 2000||Jan 13, 2004||Extreme Networks||Policy based quality of service|
|US6690649||May 20, 1999||Feb 10, 2004||Nec Corporation||QoS management apparatus|
|US6724722 *||Mar 7, 2000||Apr 20, 2004||Lucent Technologies Inc.||Managing congestion and potential traffic growth in an information network|
|US6728265||Jul 30, 1999||Apr 27, 2004||Intel Corporation||Controlling frame transmission|
|US6738349||Mar 1, 2000||May 18, 2004||Tektronix, Inc.||Non-intrusive measurement of end-to-end network properties|
|US6760314||Jun 30, 2000||Jul 6, 2004||Nec Corporation||Network load distribution system and network load distribution method|
|US6785237||Mar 31, 2000||Aug 31, 2004||Networks Associates Technology, Inc.||Method and system for passive quality of service monitoring of a network|
|US6801502||May 8, 2000||Oct 5, 2004||At&T Corp.||Method and apparatus for load-sensitive routing of long-lived packet flows|
|US6831890||Oct 31, 2000||Dec 14, 2004||Agilent Technologies, Inc.||Measuring network performance parameters in data communication networks|
|US6831893||Apr 3, 2000||Dec 14, 2004||P-Cube, Ltd.||Apparatus and method for wire-speed classification and pre-processing of data packets in a full duplex network|
|US6842783||Feb 18, 2000||Jan 11, 2005||International Business Machines Corporation||System and method for enforcing communications bandwidth based service level agreements to plurality of customers hosted on a clustered web server|
|US6857025||Apr 5, 2000||Feb 15, 2005||International Business Machines Corporation||Highly scalable system and method of regulating internet traffic to server farm to support (min,max) bandwidth usage-based service level agreements|
|US6904020||Nov 1, 2000||Jun 7, 2005||Agilent Technologies, Inc.||System and method for monitoring communication networks using data stream characterization|
|US6947379||Apr 3, 2001||Sep 20, 2005||Cisco Technology, Inc.||Packet routing to reduce susceptibility to disturbances|
|US6954431||Dec 6, 2001||Oct 11, 2005||Caspian Networks, Inc.||Micro-flow management|
|US7065482||May 17, 2001||Jun 20, 2006||International Business Machines Corporation||Internet traffic analysis tool|
|US7133365||Dec 28, 2001||Nov 7, 2006||Internap Network Services Corporation||System and method to provide routing control of information over networks|
|US7164657 *||Jan 26, 2005||Jan 16, 2007||Inmon Corporation||Intelligent collaboration across network systems|
|US7185079 *||Aug 8, 2001||Feb 27, 2007||Cisco Technology, Inc.||Automated management of network addresses in a broadband managed access environment|
|US7222190||Dec 7, 2001||May 22, 2007||Internap Network Services Corporation||System and method to provide routing control of information over data networks|
|US7269157||Apr 10, 2001||Sep 11, 2007||Internap Network Services Corporation||System and method to assure network service levels with intelligent routing|
|US20010021176||Mar 12, 2001||Sep 13, 2001||Itaru Mimura||Method of monitoring quality of communication for each flow|
|US20010037387||Mar 21, 2001||Nov 1, 2001||F5 Networks, Inc.,||Method and system for optimizing a network by independently scaling control segments and data flow|
|US20020010765||Jul 12, 2001||Jan 24, 2002||John Border||Method and system for prioritizing traffic in a network|
|US20020010792||Jul 12, 2001||Jan 24, 2002||John Border||Method and system for improving network performance by utilizing path selection, path activation, and profiles|
|US20020040400||Dec 4, 2001||Apr 4, 2002||F5 Networks, Inc.||Method and system for storing load balancing information with an HTTP cookie|
|US20020057699||Apr 19, 2000||May 16, 2002||Roberts Lawrence G.||Micro-flow management|
|US20020075813||Sep 20, 2001||Jun 20, 2002||Baldonado Omar C.||Method and apparatus for coordinating routing parameters via a back-channel communication medium|
|US20020078223||Aug 6, 2001||Jun 20, 2002||Baldonado Omar C.||Method and apparatus for performance and cost optimization in an internetwork|
|US20020103846||Feb 6, 2001||Aug 1, 2002||Radware Ltd.||Load balancing|
|US20020105909||Feb 7, 2001||Aug 8, 2002||Mark Flanagan||Quality-of-service monitor|
|US20020105911||May 23, 2001||Aug 8, 2002||Parag Pruthi||Apparatus and method for collecting and analyzing communications data|
|US20020110084||Dec 18, 2000||Aug 15, 2002||Butt Alan B.||Application based bandwidth limiting proxies|
|US20020141378||Mar 28, 2001||Oct 3, 2002||Bays Robert James||Methods, apparatuses and systems facilitating deployment, support and configuration of network routing policies|
|US20020145981||Apr 10, 2001||Oct 10, 2002||Eric Klinker||System and method to assure network service levels with intelligent routing|
|US20020163884||May 3, 2001||Nov 7, 2002||Amir Peles||Controlling traffic on links between autonomous systems|
|US20020184393||Jul 10, 2001||Dec 5, 2002||Leddy John G.||Routing information exchange|
|US20020186661||May 4, 2001||Dec 12, 2002||Terago Communications, Inc.||System and method for hierarchical policing of flows and subflows of a data stream|
|US20030002443||May 24, 2001||Jan 2, 2003||International Business Machines Corporation||System and method for enhancing the availability of routing systems through equal cost multipath|
|US20030012145||Jul 12, 2002||Jan 16, 2003||Nigel Bragg||Routing for a communications network|
|US20030016627||Jul 23, 2001||Jan 23, 2003||Melampy Patrick J.||System and method for determining flow quality statistics for real-time transport protocol data flows|
|US20030074449||Oct 12, 2001||Apr 17, 2003||Rory Smith||Bandwidth allocation in a synchronous transmission network for packet oriented signals|
|US20030076840||Oct 18, 2001||Apr 24, 2003||Priya Rajagopal||Multi-path analysis for managing machine communications in a network|
|US20030079005||May 29, 2001||Apr 24, 2003||61C Networks, Inc.||System and method for efficient wide area network routing|
|US20030086422||Dec 28, 2001||May 8, 2003||Netvmg, Inc.||System and method to provide routing control of information over networks|
|US20030088529||Nov 1, 2002||May 8, 2003||Netvmg, Inc.||Data network controller|
|US20030088671||Dec 7, 2001||May 8, 2003||Netvmg, Inc.||System and method to provide routing control of information over data networks|
|US20030118029||Sep 30, 2002||Jun 26, 2003||Maher Robert Daniel||Method and apparatus for enforcing service level agreements|
|US20030133443||Oct 29, 2002||Jul 17, 2003||Netvmg, Inc.||Passive route control of data networks|
|US20030214913||May 17, 2002||Nov 20, 2003||Chao Kan||Passive network monitoring system|
|US20040196787||Apr 16, 2004||Oct 7, 2004||Yufei Wang||Managing congestion and traffic flow by considering the minimization of link utilization values|
|US20040258226||Jul 14, 2004||Dec 23, 2004||Tierra Telecom, Inc. Aka Tierra Telecommunications||Monitoring and control module|
|US20070140128||Oct 27, 2006||Jun 21, 2007||Eric Klinker||System and method to provide routing control of information over networks|
|EP1017203A2||Dec 1, 1999||Jul 5, 2000||Northern Telecom Limited||Monitoring of Internet differentiated services for transactional applications|
|EP1063818A2||Jun 23, 2000||Dec 27, 2000||Softcom Microsystems||System for multi-layer provisioning in computer networks|
|JP2001024699A||Title not available|
|JPH1127327A||Title not available|
|WO1998058474A1||Jun 12, 1998||Dec 23, 1998||Telefonaktiebolaget Lm Ericsson||Dynamic quality control network routing|
|WO1999027684A1||Nov 23, 1998||Jun 3, 1999||Packeteer, Inc.||Method for automatically classifying traffic in a packet communications network|
|WO2002084951A1||Apr 10, 2002||Oct 24, 2002||Netvmg, Inc.||System and method to assure network service levels with intelligent routing|
|WO2003040874A2||Nov 1, 2002||May 15, 2003||Netvmg, Inc.||Providing routing control of information over networks|
|WO2003040947A1||Nov 4, 2002||May 15, 2003||Netvmg, Inc.||Data network controller|
|WO2003041342A1||Nov 1, 2002||May 15, 2003||Netvmg, Inc.||System and method to provide routing control of information over data networks|
|WO2004040423A2||Oct 29, 2003||May 13, 2004||Internap Network Services Corporation||Passive route control of data networks|
|1||International Preliminary Examination Report for PCT/US02/11357 mailed Oct. 1, 2003.|
|2||International Search Report for PCT/US02/11357 mailed Aug. 19, 2002.|
|3||International Search Report for PCT/US02/35156 mailed May 30, 2003.|
|4||International Search Report for PCT/US02/35158 mailed Dec. 23, 2002.|
|5||International Search Report for PCT/US02/35488 mailed Feb. 13, 2003.|
|6||International Search Report for PCT/US03/34446 mailed May 26, 2004.|
|7||Labovitz et al., "Delayed Internet Routing Convergence", IEEE, Jun. 2001, pp. 293-306, vol. 9, No. 3.|
|8||Official Action of Japanese Patent Office for JP 2002-582558 mailed Dec. 19, 2006.|
|9||Official Action of Japanese Patent Office for JP 2002-582558 mailed Sep. 11, 2007.|
|10||Supplementary European Search Report from the European Patent Office for EP02719487 mailed Aug. 22, 2005.|
|11||Written Opinion for PCT/US02/11357 mailed Apr. 16, 2003.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7894461 *||Nov 20, 2008||Feb 22, 2011||At&T Intellectual Property I, L.P.||Methods and apparatus to infer the status of border gateway protocol sessions|
|US8072901 *||Sep 29, 2005||Dec 6, 2011||Cisco Technology, Inc.||Technique for efficient probing to verify policy conformance|
|US8285850 *||Jan 19, 2006||Oct 9, 2012||Symantec Operating Corporation||Configuration and dynamic detection of connection-based backup policies|
|US8670359||Feb 13, 2006||Mar 11, 2014||Lg Electronics Inc.||Method of controlling data transmission for MBS in broadband wireless access system|
|US8717874||Aug 27, 2012||May 6, 2014||International Business Machines Corporation||Updating a switch software image in a distributed fabric protocol (DFP) switching network|
|US8750129||Oct 6, 2011||Jun 10, 2014||International Business Machines Corporation||Credit-based network congestion management|
|US8767529||Sep 12, 2011||Jul 1, 2014||International Business Machines Corporation||High availability distributed fabric protocol (DFP) switching network architecture|
|US8767722||Aug 27, 2012||Jul 1, 2014||International Business Machines Corporation||Data traffic handling in a distributed fabric protocol (DFP) switching network architecture|
|US8797843||Aug 27, 2012||Aug 5, 2014||International Business Machines Corporation||High availability distributed fabric protocol (DFP) switching network architecture|
|US8798080||Aug 27, 2012||Aug 5, 2014||International Business Machines Corporation||Distributed fabric protocol (DFP) switching network architecture|
|US8824485||Sep 15, 2012||Sep 2, 2014||International Business Machines Corporation||Efficient software-based private VLAN solution for distributed virtual switches|
|US8837499||May 14, 2011||Sep 16, 2014||International Business Machines Corporation||Distributed fabric protocol (DFP) switching network architecture|
|US8856801||Sep 4, 2012||Oct 7, 2014||International Business Machines Corporation||Techniques for executing normally interruptible threads in a non-preemptive manner|
|US8868735 *||Feb 2, 2012||Oct 21, 2014||Cisco Technology, Inc.||Wide area network optimization|
|US8942094||Oct 19, 2012||Jan 27, 2015||International Business Machines Corporation||Credit-based network congestion management|
|US8948003||Dec 9, 2011||Feb 3, 2015||International Business Machines Corporation||Fault tolerant communication in a TRILL network|
|US8948004||Feb 28, 2013||Feb 3, 2015||International Business Machines Corporation||Fault tolerant communication in a trill network|
|US9019950 *||May 22, 2009||Apr 28, 2015||Stmicroelectronics Rousset Sas||Data processing system having distributed processing means for using intrinsic latencies of the system|
|US9059922 *||Feb 27, 2013||Jun 16, 2015||International Business Machines Corporation||Network traffic distribution|
|US9065745 *||Oct 6, 2011||Jun 23, 2015||International Business Machines Corporation||Network traffic distribution|
|US9104986 *||Mar 9, 2009||Aug 11, 2015||Centurylink Intellectual Property Llc||Customer premise equipment with access to free market based pricing for bandwidth on a communications network|
|US9118490||Jul 5, 2012||Aug 25, 2015||Centurylink Intellectual Propert Llc||Free market based pricing for bandwidth and network usage|
|US9262093 *||Dec 16, 2013||Feb 16, 2016||Google Inc.||Prioritized rate scheduler for a storage system|
|US9276953||Aug 24, 2013||Mar 1, 2016||International Business Machines Corporation||Method and apparatus to detect and block unauthorized MAC address by virtual machine aware network switches|
|US9525638||Nov 15, 2013||Dec 20, 2016||Internap Corporation||Routing system for internet traffic|
|US20090046637 *||Feb 13, 2006||Feb 19, 2009||Yong Ho Kim||Method of Controlling Data Transmission for Mbs in Broadband Wireless Access System|
|US20100040045 *||May 22, 2009||Feb 18, 2010||Stmicroelectronics Rousset Sas||Data processing system having distributed processing means for using intrinsic latencies of the system|
|US20100124221 *||Nov 20, 2008||May 20, 2010||Dan Pei||Methods and apparatus to infer the status of border gateway protocol sessions|
|US20100226254 *||Mar 9, 2009||Sep 9, 2010||Embarq Holdings Company, Llc||Customer premise equipment with access to free market based pricing for bandwidth on a communications network|
|US20130088969 *||Oct 6, 2011||Apr 11, 2013||International Business Machines Corporation||Network traffic distribution|
|US20130182571 *||Feb 27, 2013||Jul 18, 2013||International Business Machines Corporation||Network traffic distribution|
|US20160134542 *||Nov 7, 2014||May 12, 2016||Grigore Raileanu||System and method for optimizing traffic in packet-switched networks with internet exchanges|
|U.S. Classification||709/232, 709/230|
|International Classification||H04L12/24, H04L29/06, G06F15/16, H04L12/56, H04L12/26|
|Cooperative Classification||H04L45/38, H04L43/0894, H04L43/0882, H04L43/18, H04L47/2441, H04L43/045, H04L47/20, H04L43/00, H04L45/30, H04L43/10, H04L43/0864, H04L45/00, H04L45/308, H04L43/0876, H04L43/16, H04L41/0893, H04L43/087, H04L43/028, H04L49/90, H04L29/06027, H04L43/12, H04L45/302, H04L43/0829, H04L43/022, H04L69/14, H04L69/22, H04L65/80|
|Nov 1, 2002||AS||Assignment|
Owner name: NETVMG, INC., CALIFORNIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:KLINKER, ERIC;JOHNSON, JEREMY;MCCANNE, STEVEN;AND OTHERS;REEL/FRAME:013459/0783
Effective date: 20021101
Owner name: NETVMG, INC.,CALIFORNIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:KLINKER, ERIC;JOHNSON, JEREMY;MCCANNE, STEVEN;AND OTHERS;REEL/FRAME:013459/0783
Effective date: 20021101
|Oct 20, 2009||AS||Assignment|
Owner name: INTERNAP NETWORK SERVICES CORPORATION, GEORGIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:NETVMG, INC.;REEL/FRAME:023397/0352
Effective date: 20091015
Owner name: INTERNAP NETWORK SERVICES CORPORATION,GEORGIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:NETVMG, INC.;REEL/FRAME:023397/0352
Effective date: 20091015
|Nov 9, 2010||AS||Assignment|
Owner name: WELLS FARGO CAPITAL FINANCE, LLC, AS AGENT, MASSAC
Free format text: SECURITY AGREEMENT;ASSIGNOR:INTERNAP NETWORK SERVICES CORPORATION;REEL/FRAME:025337/0437
Effective date: 20101102
|Dec 7, 2010||CC||Certificate of correction|
|Aug 23, 2013||FPAY||Fee payment|
Year of fee payment: 4
|Dec 4, 2013||AS||Assignment|
Owner name: INTERNAP NETWORK SERVICES CORPORATION, GEORGIA
Free format text: RELEASE BY SECURED PARTY;ASSIGNOR:WELLS FARGO CAPITAL FINANCE, LLC (AS AGENT);REEL/FRAME:031710/0635
Effective date: 20131126
|Dec 5, 2013||AS||Assignment|
Owner name: JEFFERIES FINANCE LLC (AS COLLATERAL AGENT), NEW Y
Free format text: SECURITY AGREEMENT;ASSIGNOR:INTERNAP NETWORK SERVICES CORPORATION;REEL/FRAME:031765/0527
Effective date: 20131126
|Feb 16, 2015||AS||Assignment|
Owner name: INTERNAP CORPORATION, GEORGIA
Free format text: CHANGE OF NAME;ASSIGNOR:INTERNAP NETWORK SERVICES CORPORATION;REEL/FRAME:035000/0349
Effective date: 20141124
|Apr 7, 2017||AS||Assignment|
Owner name: INTERNAP CORPORATION, GEORGIA
Free format text: RELEASE BY SECURED PARTY;ASSIGNOR:JEFFERIES FINANCE LLC;REEL/FRAME:041929/0328
Effective date: 20170406